Coal-fired Unit SCR Denitration System Neural Network Control Strategy Research

In recent years,due to the impact of frequent peak regulation of the power grid on the combustion system of coal-fired power plant boilers,the fluctuating concentration of nitrogen oxides(NOx)in the flue gas of coal-fired power plants has been very severe.The hysteresis of the concentration measurement system and other factors make the traditional control strategy unable to meet the control requirements of the SCR denitration system,so it is particularly important to study an advanced SCR denitration system control strategy.In order to study the advanced control strategy of a complex system,an accurate controlled object model is essential.For this purpose,this paper proposes a method for modeling the SCR denitration system in a small room for a coal-fired power plant.Based on this,this paper studies the neural network predictive control strategy and neural network inverse control strategy of the SCR denitration system,analyzes the advantages and disadvantages of these two control strategies,and finally proposes a neural network composite control strategy.The test proves that this control strategy has the advantages of the above two controllers and has a good control effect.The main content of this article is arranged as follows:1.Using LH(Langmuir-Hinshelwood)mechanism and ER(Eley-Rideal)mechanism,by analyzing the structure of the SCR denitration reactor,a mechanism modeling method for the sub-chamber SCR denitration system is proposed,and then the parameters of the proposed mechanism model are optimized.,And use this model as the controlled object to study the control strategy.2.Based on the established SCR denitration system mechanism model,construct its neural network model to study the predictive control strategy of the SCR denitration system neural network model,and conduct disturbance tests to analyze the advantages and disadvantages of this control strategy.3.Based on the established mechanism model of SCR denitration system,construct its neural network inverse model to study the neural network inverse control strategy of SCR denitration system,and conduct disturbance experiments to analyze the advantages and disadvantages of this control strategy.4.Based on the advantages of the two neural network control strategies studied earlier,this paper proposes a new neural network composite control strategy.This strategy is based on the adjustment of the output weight of the t wo controllers as the control amount based on the changes in air volume and nitrogen oxides.One method and then subject it to the same perturbation test.5.Using real field data,perform simulation tests on three neural network controllers,and compare the control effects of the three controllers with the controller output.